Hybridized structure- and ligand- based drug discovery approaches targeting ASCT2, an amino acid transporter critical for upregulated cell proliferation in numerous cancer types This proposal outlines the protocols and techniques I will be using to optimize drug discovery of ASCT2, a promising target for anti-cancer therapeutics. ASCT2 plays a key role in increasing the glutamine influx for tumor cells to maintain such high metabolic rates required for rapid proliferation. The first structures of ASCT2 were recently determined experimentally, making this a newly viable target for structure-based studies ASCT2 was only recently discovered to play a critical role in cancer cell metabolism and little medicinal chemistry efforts have been focused on ASCT2 antagonist development allowing immense potential for breaking into new compound scaffolds for further testing. Currently, there have not been any ASCT2 drug campaigns that incorporate computational drug discovery methods and this proposal outlines the first studies dedicated to this. Many institutions and pharmaceutical companies have implemented computational strategies into drug discovery pipelines as a means to produce viable drug candidates in a more cost-efficient and timely manner. Depending on the target of interest, researchers focus more intently on either ligand-based (LB) or structure-based (SB) methods, but rarely are these two methods hybridized in a sophisticated fashion. By utilizing strategies of both LB- and SB- computational drug discovery, I intend to merge the advantages of both methodologies as a means to sample and filter large chemical space more efficiently. Our lab has active development in two computational chemistry software suites: Rosetta primarily focuses on SB methods whereas the Biology and Chemistry Library (BCL) contains advanced cheminformatics toolsets for LB methods. The focus of my project will be to integrate the RosettaDrugDesign code to allow a more extensive, yet efficient sampling of chemical space using ligand-based techniques. We intend to incorporate these more advanced LB techniques available in the BCL, including multi-tasking artificial neural networks for Quantitative Structure- Activity Relationship predictions, to filter compounds during docking simulations within the RosettaDrugDesign. By bringing together the structure prediction abilities of Rosetta and small- molecule tools of BCL, we anticipate exceptional advances in our abilities to efficiently design drugs for ASCT2.